By an iterative process:
- run an all parameters optimization with long steps for each one (SL-> Init = 10, End = 100, Step = 10)
- when you get a good strategy´s performance parameters set, start with a single parameter optimization with a short step
- when you get that the reult for the single optimization parameter is better than before, keep that parameter fixed and run again a multi parameters optimization as for the first step
- go forward with a new single parameter optimization
- repeat the iterative method until you get what you think is the best parametrization
Keep in mind that the best optimized parameters set (one giving the highest profit, for example) does not mean that it is the best option to trade.

























